17 thoughts on “How Carmelo and Amare may con the voters for another year.”

1. Landry’s WP should go down playing alongside another effective defensive rebounder.

2. The narrow view of advanced statisticians is laughable to me. First of all to judge a player simply from their individual numbers is a fundamental flaw in any kind of advanced statistic because there is not any metric measuring the impact they have on other players around the floor. Want an example of this? Carmelo may not be the most efficient scorer, but along with Kobe Bryant, Dirk Nowitzki and Kevin Durant he is one of the few players in the league that forces a defense to account for them no matter where they are on the floor due to their incredible abilities to score in all manners and from any space on the floor.

   While Melo and Kobe at times leave much to be desired in terms of shot selection there is a reason that players and coaches around the league make their teams game plan for them.

   http://query.nytimes.com/gst/fullpage.html?res=9803E7D8103FF935A25752C0A9679D8B63

   In this article by Nate Silver, of Baseball and Basketball Prospectus fame, shows how in Anthony’s time with the Nuggets 16 teammates played more than 2,000 minutes with him and 2,000 mintues elsewhere in the NBA. All but two, Marcus Camby and Voshon Lenard, saw their true shooting percentages rise significantly. Allen Iverson, of poor shot selection fame, saw his true shooting percentage rise from a below average mark of 51.2% to an above average mark of 55.0% in his time with Anthony.

   Advanced metrics have their place, but you need to be able to see that there are limitations to them as well. Simply put look at the Nuggets pre-Melo team that won a grand total of 17 games and then with the addition of Carmelo made the playoffs in a highly competitive Western Conference. It is also interesting to note how efficient every offense has been that Carmelo has been the focal point of throughout his career.

   Advanced metrics would have recommended the Knicks re-signed David Lee in 2010 rather than making a run at Amar’e Stoudemire. Looking at their numbers that would seem prudent, but again a player like Amar’e commands the attention of a defense whereas most teams are content to play Lee on a man-to-man level and afford him the opportunity to put up numbers. The reason being is that Lee while an extremely effective player cannot carry the offensive load for a team, while Amar’e if left on an island with a defender can dominate a game.

   Advanced metrics would have given the Mavericks no chance playing against the kings advanced statistics in Wade and James. Thankfully champions and games won are not awarded based on numbers alone and we were able to witness how an offense built around the unique abilities of Nowitzki to score in every imaginable way on the floor and shooters to space the floor around him could defeat physical specimens that live at the rim (Not trying to sell Carlisle or Tyson Chandler short).

   Advanced metrics have definitely allowed teams, players and fans to recognize undervalued players and skill sets, but they must work in concert with using your eyes which will tell you that Carmelo and Amar’e defensive inadequacies and all are players that do have great value in the NBA.

3. I think there may be some validity to Ashwin’s statement, but then again I don’t feel like doing any research to find out. I just suspect that chuckers (aka low WP/high usage players) may actually improve the WP of players around them more so than say a Nash would.

   Regardless I don’t think Fields is going to have as productive of a season as last year. The Knicks didn’t have much in terms of a center last season, they now have a good one, so rebounds will be harder to come by. They also don’t have a PG that is as willing to shoot as Billups and Felton were. I see Fields taking more onus upon himself to score thereby lowering his FG%, and since the rebounds won’t be as plentiful he’ll also lose out on the gain there. I’m basically judging that off of one preseason game in which he took 12 shots though.

4. Michael, the new Wins Produced accounts for diminishing returns in defensive rebounds: http://wagesofwins.com/2011/12/11/wins-produced-comes-back-better-and-stronger/

   Ashwin, we commented on the Silver article when it was written: http://wagesofwins.com/2011/01/17/commenting-on-nate-silver%E2%80%99s-melo-effect/

5. Michael,
   Actually a very interesting point. The only real thing the Knicks had going for them was Fields impressive play and a lot of that is rebounding. They got Chandler, which will help but did not address the large number of inefficient shots Melo and Amare will take. While I do think Chandler was an easy choice to pick up they should also do something help on shooting as Melo and Amare won’t.

6. Ashwin,
   As James pointed out the Silver study had lots of holes in it. Another counter point is the Nuggets shot better last season after Melo left (http://thenbehteam.blogspot.com/2011/03/nuggets-knicks-and-melo-effect.html) When we look at each shot a player takes it comes at the cost of shots for their team mates. So taking inefficient shots to “help” out your team mates is not an ideal strategy.

   I will also recommend you peruse the archives. Several respected analysts including Ben Morris and Arturo Galletti put the Mavs to win against the Heat (and in Arturo’s case the Lakers)

   I will also try and give you a nudge for any future comments on advanced stats. You produced an 8 paragraph post with one link and no real stats. Next time I would love to see some actual numbers shown other than the terrible ones from an uncharacteristically bad article from Nate Silver.

7. Neutes,
   I don’t disagree. Actually I’m afraid Fields may take a bad turn. If Chandler is grabbing more boards than Fields may try less to get them. Additionally his one preseason game did have him chucking the ball up a lot. If Fields decides he’d rather get paid (by increasing PPG and focusing less on D) than help the Knicks win his production may go down.

8. Andres,

   You just stumbled upon the conundrum that using statistics to judge players produces. This is what I don’t think you guys can’t reconcile with very well – that is you make an unverified, and unverifiable, correlation between effort and WP. Fans see Melo working his tail off to score. They see Fields standing around trying to pick his spots and only taking a shot if he’s open. Who’s trying harder to help the Knicks win?

   There’s obviously just as big of an issue on the opposite end of the spectrum. Players know scoring gets them paid, but what if this wasn’t the case. What if WP got them paid? How would they adjust? Would you have players out there refusing to shoot the ball? Would they be too scared to make a mistake and mess up their WP? Would some guy end up being the sacrificial lamb that the rest of his teammates decided to stick him with the ball at the end of every shot clock forcing him to chuck up 20 contested shots a game?

9. I’m not disputing the usefulness of advanced statistics. They have their uses and the fact that their is a clear link between NBA franchises that employ advanced statisticians and success versus those that don’t indicates how important they have become in the decision making process.

   All I was trying to point out is that their are limitations to its usefulness and that every metric is subject to the bias of the person that creates it. Thus their are differences between John Hollinger’s PER, Basketball Prospectus’ WARP and Wages of Wins’ own WP and WP/48. These are all useful metrics that can help to inform people of where a player ranks.

   I do believe that as much as good individual players are often are stuck on bad teams that often times they can also benefit statistically from that. Not to beat a dead horse, but David Lee is a player who in the summer of 2010 benefited greatly from that (in my opinion) based on becoming a double-double machine during the 2009-10 season.
   https://spreadsheets.google.com/pub?key=0AsCknWukkAO3dEQ4bVZ3WlAwQkgyRmE2cW1QcndWUXc&hl=en&gid=0
   According to this chart produced by this website David Lee was more productive than Dirk Nowitzki, Paul Pierce, and Chris Bosh.

   My question based on looking at this is, is team success a factor in your WP48 or is it simply an individual metric? Is the pace at which a player’s team plays at already factored in to WP48?

   I also have a random unrelated question. During the season and playoffs last year when watching the Mavericks play, their personnel, the way that personnel was used both offensively and defensively and Rick Carlisle’s rotations and constant tinkering I had a feeling that they were using statistics consistently to inform their choices and decisions. I’m not sure if you felt that way as well or if I’m simply reacting to what seemed to be a team that was greater than the sum of its parts, at least at first glance. Thanks for providing the response to Silver’s article and I do enjoy reading the work you all do here.

10. IMO, Amare’s efficiency declined last year because he wasn’t playing with Nash anymore. Nash got him a lot of high quality “assisted FGs” at the rim. For the Knicks Amare had to create more shots for himself and wound up shooting more mid range jumpers or putting the ball on the floor trying to get to the hoop on his own . The latter lead to more turnovers also.

    The Knicks still don’t have a play making PG anywhere near the quality of Nash. So that issue remains, but if Baron Davis gets healthy and focuses on play making instead of bad shot selection, he might be able to help Amare.

    Keep in mind that Chandler’s high scoring efficiency has benefited from playing with CP3 and Kidd feeding him alley oops. His efficiency might also be impacted on a team without a true play maker unless D’Antoni can make some adjustments.

11. IMO David Lee’s drop in productivity last year is not a great way to try to evaluate the validity of this metric (at least not yet).

    In his last year for the Knicks, D’Antoni made Lee a focal point of the offense. The Knicks ran lot of P&Rs for him and the offense actually ran through him a lot of the time. In Golden State, Monta Ellis and Curry take a huge number of shots. So Lee’s role was both reduced AND changed. His usage went down and he was taking more short and mid range jumpers instead (which he failed to make at highly effective rates). I think you could argue that Lee’s productivity declined because Golden State didn’t use him the same way
    (properly).

    In addition, Lee was playing hurt for a good chunk of the season. So at the margin, that may have impacted him also.

    Perhaps Mark Jackson will use him differently, but as long as Monta Ellis is a Warrior, you can expect Lee’s usage to be lower and also expect him to get fewer easy baskets.

12. Andres,

    There is one thing about usage, ball handling, and turnovers that bothers me.

    It seems to me that the higher a player’s ball handling and usage goes, the higher his turnovers tend to be. To offset all those extra TOs a high usage ball handler has to do a lot of other good things to make up for those TOs.

    For a play making PG, it doesn’t seem to be much of an issue because they often accumulate more than twice as many assists as turnovers. So all that extra ball handling nets out to a positive anyway (if they are good).

    For more pure scorers (like Melo), handling the ball a lot leads to the extra turnovers without enough of a positive statistical offset. So they wind up getting rated lower.

    The thing is, part of Melo’s role on the team is to score in isolation when the defense isn’t giving the team anything better. So he’s getting punished for being the most effective player on the team at scoring on the bad possessions.

    If The Knicks put Fields in that role, his turnovers would skyrocket and his efficiency plummet because he can’t handle the ball to save his life (and can’t score off the dribble well).

    I think it’s possible that non PG ball handlers are getting unfairly punished for the extra turnovers that come with having more skill without necessarily having the full play making skills of a PG . Hope that’s clear.

13. Neutes:

    “Fans see Melo working his tail off to score. They see Fields standing around trying to pick his spots and only taking a shot if he’s open. Who’s trying harder to help the Knicks win?”

    There is just so much fail in this statement. The simple truth is that people make loads of excuses for players they like or that they “know are good”. One of my favorites is Bruce Bowen. WP48 always very consistently said that Bowen was terrible, pundits said “he’s a champion, plays good defense that you cannot measuer”, etc yadda yadda yadda.

    Then, however, when WP48 points to players like Landry Fields or Ronnie Brewer, and says that they are very good despite the fact that they don’t score, it’s the other way around, despite the fact that those players are defensive “role players”.

    I genuinely have little patience with those of you who just “know that you are right” and who choose to post about it in the comments. If that’s the case I guess we’ll all come back here in a couple of months. I guarantee that one of two things will happen:

    a) New York will be “meh” and you’ll all blame it on chemistry rather than admit that perhaps WP48 measures *something*
    b) New York will be having a great year and you’ll all be rubbing our noses in how WP48 is worthless because it doesn’t measure “effort”, “intangibles”, or “shot creation”.

    Notice the asymetrical nature of those two outcomes?

14. lovethoseknicks,

    WP48 is a per-minute (well, per 48 minutes) stat. Why should a player that handles the ball a lot turn over the ball on a per-minute basis more than a player who rarely handles it? That makes no sense.

    It makes some sense that ball handlers will have higher turnover rates than big men, because it’s easier for the defense to apply pressure to them. But it makes zero sense that Carmelo’s turnovers *per 48 minutes* are any higher than an another small forward’s turnovers *per 48 minutes*.

    Also, I’d be willing to bet a large sum of money that amongst players who play the same position, usage rate and turnovers-per-48 do not highly positively correlate (i.e. if you are going to make the argument that Melo makes more turnovers because the team always gives him the ball in pressure situations, I do not buy that). In fact, they *should* negatively correlate — it would seem to me that if you have two small forwards, in any given high pressure situation, it would be a good strategy to give the ball more often to the guy who is less likely to turn it over.

15. TheNBAGeek,

    “It makes some sense that ball handlers will have higher turnover rates than big men, because it’s easier for the defense to apply pressure to them. But it makes zero sense that Carmelo’s turnovers *per 48 minutes* are any higher than another small forward’s turnovers *per 48 minutes*.”

    I don’t think I was clear.

    I believe that Melo handles the ball more often than the average SF per 48 minutes.

    When you handle the ball, there are 3 primary statistical outcomes.

    1. Shoot, get fouled etc… (make or miss)
    2. Pass and get an assist
    3. Turn the ball over

    I looked at the league average a couple of years ago. I think it was approximately 3 assists for every 2 turnovers per 36 minutes.

    Valuing assists at .5 and TOs at -1

    1.5 – 2 = -.5

    So if we conclude that handling the ball more often leads to both more assists and turnovers, the more you handle it the more negative you are going to be “on average”. That’s why the great PGs handle it so much. Their handling and passing skills are so high their assist to TO ratio is high enough to produce positive value, most players can’t.

    I think you could argue that Melo SHOULD NOT handle the ball so much, but I don’t think that was entirely a choice. It has often been his job to get shots and make plays in isolation.

    This year, since the Knicks don’t have true starting PG, they are actually planning to run the offense through Melo a lot of the time (at least until Baron Davis is healthy) . So it’s quite possible his assists and TOs will hit an all time high (or close).

    I saw that happen with Amare last year. Both his assists and TOs were up last year because he had to handle the ball more often in NY than he had to on the Suns.

    I’m not even sure what I am saying here other than something doesn’t make sense.

16. Ho jbrocato, thanks for the link, I didn’t realise WP had been improved. Are there any plans to write posts looking at the historical greats (Jordan, Bird etc) using the new system?

17. Field productivity will be going way down, not his per minute producutivity, but his total. Less time , less touches etc. IT happened last year after carmelo came. He won’t even touch the ball this year, and will see a lot less time. They dont think he is any good. And now they have baron davis to hog the ball too. He will wind up somewhere like houston